Äîêóìåíò âçÿò èç êýøà ïîèñêîâîé ìàøèíû. Àäðåñ îðèãèíàëüíîãî äîêóìåíòà : http://sn.sai.msu.ru/~sil/preprints/n5055.ps.gz Äàòà èçìåíåíèÿ: Mon Jun 3 19:17:56 2002 Äàòà èíäåêñèðîâàíèÿ: Mon Oct 1 19:39:56 2012 Êîäèðîâêà: Ïîèñêîâûå ñëîâà: ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ï ð ï ð ï ð ï ð ï ð ï ð ï ð ï ð ï ð ï

A&A 388, 461--469 (2002)
DOI: 10.1051/0004­6361:20020571
c
# ESO 2002
Astronomy
&
Astrophysics
A lopsided chemically distinct nucleus in NGC 5055 #
V. L. Afanasiev 1 and O. K. Sil'chenko 2,3
1 Special Astrophysical Observatory, Nizhnij Arkhyz, 369167 Russia
2 Sternberg Astronomical Institute, University Av. 13, Moscow 119992, Russia
3 Isaac Newton Institute of Chile, Moscow Branch
Received 23 November 2001/ Accepted 12 April 2002
Abstract. Kinematics and stellar population properties in the center of nearby Sbc galaxy NGC 5055 are studied
with the Multi­Pupil Spectrograph of the 6 m telescope of the Special Astrophysical Observatory of Russian
Academy of Sciences (SAO RAS). We confirm the rotation and stellar velocity dispersion asymmetries along the
major axis reported earlier by other authors. We have found a resolved chemically distinct core in NGC 5055, with
the magnesium­enhanced region shifted by 2. ## 5 (100 pc) to the south­west from a photometric center, toward a
kinematically identified circumnuclear stellar disk. Mean ages of stellar populations in the true nucleus, defined as
the photometric center, and in the magnesium­enhanced substructure are coincident and equal to 3--4 Gyr being
younger by several Gyr with respect to the bulge stellar population. A possible origin of the asymmetries in the
center of NGC 5055 is discussed.
Key words. galaxies: individual: NGC 5055 -- galaxies: nuclei -- galaxies: stellar content --
galaxies: kinematics and dynamics -- galaxies: evolution
1. Introduction
Very nearby luminous galaxies seem to be studied in detail
because each of them has a long list of papers devoted to
various aspects of its appearance. But a greater amount
of information always reveals an extreme complexity and
individuality of the target, so no galaxy is understood in
detail. We will use NGC 5055 to illustrate this idea. The
main characteristics of this galaxy are given in Table 1.
A large nearby Sbc galaxy, NGC 5055 attracted the
attention of investigators from the beginning of the quan­
titative extragalactic researches. Fish (1961) did photo­
graphic surface photometry of the galaxy, and Burbidge
et al. (1960) obtained a rotation curve of the ionized gas
with the slit aligned with the major axis of the galac­
tic isophotes. Later, Bosma (1981) studied the rotation
and structure of its neutral hydrogen disk and found that
the H I disk was twice as extended as the stellar one
and showed a noticeable warp beyond the boundaries of
the optical image. Afterward kinematical and photometric
Send o#print requests to: O. K. Sil'chenko,
e­mail: olga@sai.msu.su
# Based on observations collected with the 6 m telescope of
the Special Astrophysical Observatory (SAO) of the Russian
Academy of Sciences (RAS) which is operated under the fi­
nancial support of Science Department of Russia (registration
number 01­43) and on data from the HST Archive.
Table 1. Global parameters of NGC 5055.
NGC 5055
Type (NED 1 ) SA(rs)bc
R25 , kpc (LEDA 2 ) 17.2
B 0
T (RC3 3 ) 9.03
MB (LEDA) --21.27
B - V (RC3) 0.72
V r (radio, LEDA), km s -1 503
Distance (LEDA, H0 = 75 km s -1 Mpc -1 ) 9.2 Mpc
Inclination (LEDA) 55.5 #
PAphot (LEDA, NED) 105 #
vm , km s -1 (LEDA) 211
## , km s -1 (Heraudeau & Simien 1998) 117 ± 7
1 NASA/IPAC Extragalactic Database.
2 Lyon­Meudon Extragalactic Database.
3 Third Reference Catalogue of Bright Galaxies.
studies of NGC 5055 were repeated more than once, and
often the results were controversial.
Firstly, the spiral arm classification of NGC 5055 re­
mains ambiguous. In the optical Elmegreen (1981) and
Elmegreen & Elmegreen (1987) classified NGC 5055 as a
flocculent spiral galaxy. But in the NIR (K # )­ and CO­
images Thornley (1996) and Thornley & Mundy (1997)

462 V. L. Afanasiev and O. K. Sil'chenko: NGC 5055
saw two disconnected grand­design spiral patterns, inner
and outer, with a boundary at 1.5--2.0 kpc from the center.
Secondly, though the classification of NGC 5055 as Sbc
is certain, we do not know yet if the galaxy is disk­ or
bulge­dominated. Thornley (1996), by decomposing the
K # ­brightness profile into an exponential disk and a de
Vaucouleurs' bulge, found that the bulge is almost ev­
erywhere more luminous than the disk, including the re­
gions with grand­design stellar spiral arms. Fillmore et al.
(1986) found that the disk of NGC 5055 dominates the
bulge at R # 7 ## , but according to decomposition results
by Baggett et al. (1998), the disk starts to dominate at
R = 35 ## . Determinations of the disk exponential scale­
length range from 40 ## (Acaretta et al. 1996) to 98 ## (Kent
1987). Our experience of brightness profile decomposi­
tions with NGC 7331 (Sil'chenko 1999b) and NGC 7217
(Sil'chenko & Afanasiev 2000) shows that when such a va­
riety of decomposition results exists, a global disk may be
readily divided into several decoupled segments. Two dis­
connected parts of the spiral pattern in NGC 5055 make
this feasible.
The results of kinematical studies are also unusual.
Fillmore et al. (1986) obtained long­slit cross­sections
along the major and minor axes and compared the line­
of­sight velocity distributions for the stars and ionized gas
with the model predictions based on the brightness distri­
bution. Both observed velocity profiles, and especially the
one for the stars, appeared to be strongly asymmetric.
The relative (rotation?) velocities of the stars to the east
of the nucleus are twice as low as those to the west, and the
corresponding stellar velocity dispersions di#er in the op­
posite sense. Fillmore et al. (1986) note that the isophote
shape in the center of NGC 5055 implies strong dust influ­
ence to the west and south­west of the nucleus; perhaps, on
one (the eastern) side they saw the bulge and on the other
(western) side the (warped) disk. However, simultaneously
they detected ``weaker absorption lines'' just in the region
where the kinematics implied the bulge; this combination
seemed to them improbable. So the puzzle remains to be
solved. Pismis et al. (1995) analysed ionized­gas line­of­
sight velocity distributions along four di#erent directions
obtained under good seeing conditions (1 ## -1. ## 3), and they
also found a noticeable asymmetry along the major axis:
within R # 3 ## to the west of the nucleus there is an excess
of relative gas velocity. The authors treat this excess as a
manifestation of gas radial outflow due to mild nuclear
activity -- a rather strange interpretation considering that
on the major axis the projection of radial velocities onto
the line of sight is zero.
Among the other interesting features of NGC 5055 we
must mention its ultraviolet nucleus. The galaxy is classi­
fied as a ``UV­bright LINER'' -- within the heterogeneous
class of LINERs, the ``UV­bright'' ones are usually thought
to possess a non­thermal ionizing source. But in NGC 5055
the ultraviolet nucleus mapped by HST/FOC with high
spatial resolution has appeared to be resolved: its diameter
is 6 h -1 pc and its absolute magnitude is MB,nuc = -11.2
Table 2. 2D spectroscopy of NGC 5055.
Date Exposure Seeing Sp. range PA(top)
14/15.06.99 40 min 1. ## 5 4250--5600 š A 299. # 3
15/16.06.99 45 min 1. ## 2 5850--7200 š A 293. # 7
(Maoz et al. 1995), which implies that the UV nucleus of
NGC 5055 may be a young stellar cluster.
Some years ago we (Sil'chenko 1994) compiled a list
of galaxies that were candidates for possessing chemically
distinct nuclei based on redder colours of the nuclei with
respect to the bulges. NGC 5055 was not included in this
list. But later we saw from the photometric survey of
Heraudeau & Simien (1996) that it has a very red colour,
V - I > 1.5, of the central region within R = 10 ## . To
check if this is a dust concentration or a chemically dis­
tinct nucleus, we needed panoramic spectral observations.
We have observed NGC 5055 with the Multi­Pupil Fiber
Spectrograph (MPFS) of the 6 m telescope in the frame­
work of our program of searching for chemically distinct
nuclei in spiral galaxies, and we found such a nucleus in
this galaxy.
2. Observations and data reduction
In 1999 we undertook two­dimensional spectroscopy of
NGC 5055 with the Multi­Pupil Fiber Spectrograph
(MPFS) of the 6 m telescope of the Special Astrophysical
Observatory (Nizhnij Arkhyz, Russia). Exposures were
obtained in two spectral ranges: 4250--5600 š A (blue­green),
and 5850--7200 š A (red). Detailed information for the ob­
servations is given in Table 2. A grating of 1200 grooves
per mm was used that provided a reciprocal dispersion of
1.35 š A per pixel and a spectral resolution of 4 š A. A value
for the seeing FWHM = 1. ## 2-1. ## 5 was estimated from an
exposure of a double star, STF 1947, which was used to
calibrate the orientation of our pupil frame on the sky.
These spectral observations were made with the new
version of the panoramic spectrophotometer which be­
came operational at the prime focus of the 6 m telescope
at the end of 1997. With respect to the previous vari­
ants of MPFS (Afanasiev et al. 1990, 1996), the field of
view is now increased and the common spectral range is
larger due to the use of fibers: they transmit light from
16 â 15 square elements of the galaxy image to the slit of
the spectrograph (240 fibers) together with the sky back­
ground taken 4. # 5 away from the galaxy itself (16 fibers).
The size of one spatial element is 1 ## â 1 ## . At the exit of
the spectrograph a 1024â1024 CCD registers all 256 spec­
tra simultaneously. The primary reduction of the data is
made within IDL. After bias subtracting, flatfielding, and
one­dimensional spectra extraction from the CCD frame,
we linearize and analyse each spectrum individually. The
one­element spectral characteristics, such as flux in the
continuum or in emission lines, redshift, and absorption­
line indices are then combined into two­dimensional ar­
rays corresponding to the galactic region under consid­
eration using software developed earlier in the Special

V. L. Afanasiev and O. K. Sil'chenko: NGC 5055 463
Astrophysical Observatory (Vlasyuk 1993) and our own
programs. To calculate absorption­line indices and their
errors we also used the program of Dr. Vazdekis. We ob­
tained two­dimensional surface brightness distributions,
velocity fields, and maps of stellar population character­
istics. In the blue­green spectral range, we measured the
absorption­line indices H#, Mgb, Fe5270, and Fe5335 in
the Lick system (Worthey et al. 1994); to check the con­
sistency of our measurements with the model indices cal­
culated in this system (Worthey 1994), we regularly ob­
served stars from their list (Worthey et al. 1994). The
duration of exposure in the blue­green was chosen to be
long enough to provide signal­to­noise ratios of about 100
(per Angstrom) in the nucleus and #20 near the edges of
the frames; the corresponding random error estimations
made following Cardiel et al. (1998) range from 0.15 š A in
the center to 0.6--0.8 š A for the individual spatial elements
in the outer part. To keep a constant level of accuracy
along the radius, we co­added the spectra in concentric
rings centered on the nucleus, traced the radial variations
of the azimuthally­averaged absorption­line indices, and
compared them to the synthetic models of old stellar pop­
ulations of Worthey (1994) and Tantalo et al. (1998). We
estimate the mean accuracy of our azimuthally­averaged
indices as 0.1 š A. Besides index mapping, we use our blue­
green spectra to derive a stellar velocity field in the cen­
ter of NGC 5055 by cross­correlating one­element galactic
spectra with the spectra of three K­giant stars with known
line­of­sight velocities. In the red spectral range we mea­
sured baricentric positions of the emission line [N II]#6583,
which is the strongest in the center of NGC 5055, and also
of the H# emission line, to derive the velocity field of the
ionized gas. We estimated the best accuracy of our ve­
locity measurements as 10 km s -1 from the night­sky line
[O I]#6300 analysis.
In addition to the 2D spectral data, we used some
archive photometric data. NGC 5055 was observed by
HST, in particular with NICMOS--CAM3 (ID 7919, PI
W. Sparks). The galaxy was exposed on June 4, 1998,
during 192 s through a F160W (H­continuum) filter and
during 512 s through a F187N (Pa#) filter. The results of
these observations are reported in detail by Boker et al.
(1999); we use them to estimate morphological parame­
ters of isophotes. To extend our photometric analysis to
the outer parts of the global galactic disk, we also took
recourse to ground­based photometry; we used the public
data of Frei et al. (1996), obtained at the Palomar 1.5 m
telescope through Thuan & Gunn's gri­filters: though of
medium resolution (seeing of 2. ## 1-2. ## 7), these data are
deep and well­calibrated.
3. Stellar population and ionized gas in the center
of NGC 5055
To study stellar population properties in the center of
NGC 5055, we use the Lick indices H#, Mgb, and #Fe# #
(Fe5270+Fe5335)/2. Models of simple stellar populations,
e.g. by Worthey (1994), make it possible to determine si­
multaneouly a luminosity­weighted mean age and a mean
metallicity for the stellar populations by comparing H#
with Mgb or H# with #Fe#, because the Balmer absorp­
tion lines are more sensitive to the age and the metal­line
indices are more sensitive to the metallicity.
Figure 1 presents 2D distributions of the Lick indices
mentioned above in the center of the galaxy. For a chem­
ically distinct nucleus, both metal­line indices (or some­
times only Mgb), have to peak in the photometric cen­
ter of a galaxy. Indeed, both #Fe# and Mgb vary over the
field of view showing maximum values near the center.
But whereas the #Fe# peak coincides with the photomet­
ric center of the galaxy, the Mgb­enhanced area is shifted
to the south­west from the nucleus; it is well resolved,
with its own center #2. ## 5 from the photometric center of
the galaxy. The whole Mgb­enhanced structure looks like
one half of a nuclear bar or like one half of a highly in­
clined circumnuclear disk. Its contrast over the surround­
ing bulge region, #Mgb of 0.6--1.0 š A, exceeds 3#. The
#Fe#­enhanced area is more symmetric around the center
though a weak ``plume'' toward the Mgb maximum can be
noted. Earlier we found di#erent Mg and Fe distributions
in the centers of galaxies possessing chemically distinct
nuclei, but they were di#erent in opposite sense -- usually,
the Mgb peak was unresolved and coincided with the op­
tical nucleus, and the iron­enhanced areas were more ex­
tended and were treated by us as ``Fe­rich circumnuclear
disks'' (see e.g. NGC 1023, Sil'chenko 1999a; NGC 7331,
Sil'chenko 1999b; or NGC 4594, Emsellem et al. 1996).
The distribution of H#, though contaminated by emission
at R > 3 ## , can be used for stellar population analysis
in the center; it matches qualitatively the distribution of
#Fe#: the enhanced H# absorption peaks in the photomet­
ric center of the galaxy. This coincidence is explicable: to
obtain a higher iron abundance in stars, the star forma­
tion burst has to be long enough to allow the supply of
a large amount of iron from SNeIa, and such a long star
formation burst would decrease the mean age of the in­
tegrated stellar population, thus increasing the H# index.
But to determine correctly both age and metallicity, we
must use ``H# vs. metal index'' diagrams.
Firstly we must assure ourselves about the magnesium­
to­iron ratio in the central stellar population of NGC 5055.
If this ratio is not solar, and we apply models with
the solar Mg/Fe to these data, various metal­line in­
dices, when confronted with H#, would give di#erent age
estimates. Figure 2 (top) presents a ``#Fe# vs. Mgb''--
diagram to provide the necessary analysis. As was shown
by Worthey et al. (1992), in the ``iron index vs. magne­
sium index''--diagrams the models of stellar populations
with solar Mg/Fe ratio are concentrated within a nar­
row locus independent of their ages or initial stellar mass
function. Any deviation from this locus signifies a non­
solar magnesium­to­iron ratio. Worthey et al. (1992) found
that the majority of elliptical galaxies lie to the right of
the model locus, being mostly magnesium overabundant.
Theoreticians now explain this feature by a main star

464 V. L. Afanasiev and O. K. Sil'chenko: NGC 5055
Fig. 1. Lick indices maps (gray­scaled) for the central part of NGC 5055 overlaid by the green continuum isophotes; < Fe >#
(Fe5270+Fe5335)/2. The maps are spatially smoothed by a 2D Gaussian with # = 0. ## 75 except the central 5 ## â 5 ## .
formation epoch with duration less than 1 Gyr which fin­
ished before the bulk of SNeIa exploded. In disk galax­
ies, the statistics of the magnesium­to­iron ratio is more
heterogeneous due perhaps to a larger variety in their cir­
cumnuclear evolution. In NGC 5055 (Fig. 2, top) the first
impression is that the magnesium­to­iron ratio is close to
the solar one and does not change substantially along the
radius. But when we look at the next diagram, ``H# vs.
#Fe#'', or at the diagram of ``H# vs. [MgFe]'' (Fig. 2 middle
and bottom), we discover that because of the high H# in­
dex in the photometric center of the galaxy the mean age
of the nuclear stellar population cannot be larger than
3--4 Gyr. This means that the position of the galactic nu­
cleus, and also of the nearest circumnuclear outskirts, in
the diagram ``#Fe# vs. Mgb'' suggests rather a mild mag­
nesium overabundance if compared to the 5­Gyr model
sequence. Therefore, to quantify carefully the population
parameters we must also involve the models of Tantalo
et al. (1998) calculated for [Mg/Fe] = +0.3. The mid­
dle and bottom parts of Fig. 2 present age­diagnostic di­
agrams for [Mg/Fe] = +0.3 (the models of Tantalo et al.
1998) and for [Mg/Fe] = 0 (the models of Worthey 1994);
the right value of the age lies between the estimates made
from these two diagrams. To use the absorption­line in­
dex H# for the stellar population diagnostics, we must
correct it for the emission contamination which is not
negligible at R > 3 ## . We have done it in the follow­
ing way: we have summed the red spectra in the same
concentric rings as the green ones, have calculated the
equivalent widths of the H# emission line, EWH# (R), and
then have estimated the correction for the H# emission as
EWH#,em = 0.25 EWH#,em (Stasinska & Sodre 2001). The
values of H# indices in Fig. 2 have all been corrected for
the emission. By inspecting the Fig. 2, middle and bottom,
we conclude that the mean ages of the stellar populations
in the nucleus and in the magnesium­enhanced region to
the SW of the nucleus are roughly the same and are cer­
tainly less than 5 Gyr; we estimate them as 3--4 Gyr. The
mean metallicity of the nucleus is higher than the solar,
[Fe/H] nuc # +0.2-+ 0.3, and the mean metallicity of the
SW magnesium­rich ``island'' is higher by 0.2--0.3 dex than
that of the nucleus. Farther from the center, the mean age
of the stellar population rises sharply and the Mg/Fe ra­
tio approaches the solar one. Basing ourselves mainly on
the models of Worthey (1994) (Fig. 2, bottom) we esti­
mate the mean age of the stellar population in the bulge
of NGC 5055, at R = 2 ## -5 ## , as 8--10 Gyr.
To illustrate how severely the H# index is con­
taminated by Balmer emission in the central region of
NGC 5055 and to give an impression of the ionized gas dis­
tribution and excitation, we present surface brightness dis­
tributions of the red emission line intensities in Fig. 3. The
H# emission is negligible in the nucleus (see also Pogge's,
1989, statement that NGC 5055 lacks emission lines in the
nucleus), but there are several bright spots, including one
on the major axis at #5 ## to the east of the nucleus and
one at 9 ## to the south­west of the nucleus. The latter H II
region was earlier detected and noted in their Conclusions
by Pismis et al. (1995). Therefore, though NGC 5055 is
known as a LINER, it also possesses ``hot spots'' -- sites of
intense star formation? -- in its circumnuclear area. The in­
tensity distribution of [N II]#6583 is peaked in the nucleus
as expected for the LINER. However, this distribution is
noticeably asymmetric around the center, the eastern part
being brighter. If we remember the asymmetry of the stel­
lar kinematics along the major axis of NGC 5055 reported
by Fillmore et al. (1986) and its interpretation as bulge
obscuration by dust to the west of the nucleus, we have
to conclude that the [N II] emission is probably related
mostly to the bulge. Besides, the doubt of Fillmore et al.
(1986) that the weaker absorption lines to the east of the
nucleus contradict the dynamical arguments for the bulge

V. L. Afanasiev and O. K. Sil'chenko: NGC 5055 465
Fig. 2. ``Index­index'' diagnostic diagrams for the azimuthally
averaged Lick indices in the center of NGC 5055 taken along
the radius in steps of 1 ## (open circles); (top) < Fe > vs. Mgb
diagram, with the models of Worthey (1994) for [Mg/Fe] = 0,
(middle) H# vs. < Fe >, with the models of Tantalo et al.
(1998) for [Mg/Fe] = +0.3, and (bottom) H# vs. [MgFe] #
(Mgb < Fe >) 1/2 , with the models of Worthey (1994) for
[Mg/Fe] = 0. In addition to the azimuthally averaged data we
have also plotted the Mg­rich compact region to the south­west
of the nucleus (``SW­island'') by a black dot. Small signs con­
nected by thin lines present stellar population models of equal
ages; the metallicities for the Worthey's models are +0.50,
+0.25, 0.00, --0.22, --0.50, --1.00,--1.50, --2.00, if one takes the
signs from the right to the left, and for the models of Tantalo
et al. they are +0.4, 0.0, and --0.7.
visibility here now can be put away because of the detec­
tion of the chemically distinct nucleus in the form of an
asymmetric circumnuclear disk­like structure.
4. Kinematics of the stars and ionized gas
in NGC 5055
The 2D spectroscopy provides us with full line­of­sight ve­
locity fields for both stars and ionized gas that are much
more informative than long­slit cross­sections. Figure 4
presents isovelocities of the ionized gas mapped by mea­
suring independently the H# and [N II]#6583 emission
lines. The velocity field looks rather regular and reflects
mostly plane gas rotation; a faster rotation implied by the
H# measurements may be an artifact of the strong stellar
H# absorption line and slower stellar rotation with respect
to the ionized gas. Figure 5 gives the distributions of the
stellar line­of­sight velocities and stellar velocity disper­
sion. Indeed, stars rotate slightly slower than the ionized
gas. The disturbance of the stellar velocity field to the
south­west of the center seems to be stronger than the
similar isovelocity twisting in Fig. 4, but in general we
may conclude that the velocity anomaly at R # 3 ## to the
west and south­west of the nucleus which was reported
by Pismis et al. (1995) for the ionized gas, is found to be
present in the velocity distributions of both the stars and
the ionized gas. The stellar velocity dispersion distribu­
tion (Fig. 5, right) also looks asymmetric: there is a local
minimum of the velocity dispersion in the nucleus, and its
maximum area has an arc­like shape and is prominent to
the west of the photometric center. We find this rather
puzzling.
Whereas long­slit kinematical techniques can in
general provide information about real motions of stars
and gas only on a priori assumptions, usually on the
assumption of circular (axisymmetric) rotation, the
2D velocity fields give a possibility to diagnose the
character of these motions, in particular, to verify the
validity of the circular rotation paradigm. If we have an
axisymmetric mass distribution, and rotation on circular
orbits, the direction of maximum central line­of­sight
velocity gradient (we shall call it ``kinematical major
axis'') should coincide with the line of nodes as well as
the photometric major axis; whereas in the case of a
triaxial potential the isovelocities align with the principal
axis of the ellipsoid, and generally the dynamical and
photometrical major axes diverge, turning in opposite
senses with respect to the line of nodes (e.g. Monnet
et al. 1992; Moiseev & Mustsevoy 2000). In the simple
case of cylindric (disk­like) rotation we have a convenient
analytic expression for the azimuthal dependence of the
central line­of­sight velocity gradient within the area of
solid­body rotation:
dv r /dr = # sin i cos(PA - PA 0 ),
where # is the deprojected central rotation angular ve­
locity, i is the inclination of the rotation plane, and PA 0

466 V. L. Afanasiev and O. K. Sil'chenko: NGC 5055
Fig. 3. Emission­line intensity maps (gray­
scaled) for the central part of NGC 5055 over­
laid by the red continuum isophotes.
Fig. 4. Ionized­gas line­of­sight velocity fields
in the central part of NGC 5055 (isolines) as
measured by the H# and N[II]#6583 emission­
line baricenter positions. The gray­scaled back­
ground presents the red continuum intensity.
Fig. 5. The line­of­sight velocity field of the
stellar component (left, isolines) and the stellar
velocity dispersion map (right, gray­scaled) in
the central part of NGC 5055. The continuum
intensity is shown as gray­scaled on the left
plot and as isolines on the right plot.
is the orientation of the line of nodes, coinciding in the
case of the axisymmetric ellipsoid (or a thin disk) with
the photometric major axis. So by fitting azimuthal vari­
ations of the central line­of­sight velocity gradients with a
cosine curve, we can determine the orientation of the kine­
matical major axis by its phase and the central rotation
angular velocity by its amplitude.
Figure 6 presents the azimuthal variations of the line­
of­sight velocity gradients in a radius ranging from 1. ## 5 to
4. ## 0 for the stars and for the ionized gas, which are traced
by H# and [N II]#6583 emission lines; Table 3 contains
the parameter values obtained by fitting these measure­
ments by a cosine function in somewhat narrower radial
bins. First of all, one can see from Table 3 that due to the
good seeing quality we detect a certain decrease of the
gas­rotation angular velocity over a range of 1 ## -5 ## in ra­
dius. We do therefore not confirm the result of Pismis et al.
(1995) concerning solid­body gas rotation of NGC 5055
up to the radius of 3 ## . As the solid­body rotation area
is traced by us to at most R = 1. ## 5 (and actually it is
even smaller), the formula for the azimuthal velocity gra­
dient variations given above ceases to be precise, but still
remains approximately valid, because due to small ellip­
ticity of isophotes in the center of NGC 5055 (Fig. 7) the
projected radius is close to the true one. Indeed, the mea­
surements in Fig. 6 are well fitted by a cosinusoid. We just
note one feature of the azimuthal dependencies of Fig. 6: in
the position angle range of 230 # -310 # all the plots demon­
strate weaker or stronger deviations from the cosinusoids,
in other words, from circular rotation in the sense that

V. L. Afanasiev and O. K. Sil'chenko: NGC 5055 467
Table 3. Parameters of the azimuthal velocity­gradient variations fitting.
Component Radius range of fitting PA0 # sin i, km s -1 arcsec -1
Ionized gas 0. ## 9-1. ## 3 100 # ± 1 # 65 ± 9
Stars 1. ## 4-2. ## 3 97 # ± ... 22 ± 6
Ionized gas 1. ## 9-2. ## 7 104 # ± 3 # 31 ± 9
Stars 2. ## 4-3. ## 7 95 # ± ... 15 ± 4
Ionized gas 2. ## 8-4. ## 0 103.5 # ± 0.5 # 25 ± 8
Ionized gas 3. ## 8-5. ## 3 100 # ± 7 # 21 ± 6
Fig. 6. The azimuthal dependencies of the line­of­sight velocity
gradients for the stars (top) and for the ionized gas (middle and
bottom) in the center of NGC 5055 fitted by cosine curves with
a least­square method. Note the deviations from the cosine
law (or from circular flat rotation) at the line of nodes. For the
stars, black dots show the measurements in the R = 1. ## 4-2 ## 3
range, and open circles represent the measurements in the R =
2. ## 4-3 ## 7 range; a dashed­line cosine curve fitting formally the
latter data demonstrates a lower rotation­velocity amplitude --
but is it a real e#ect? The maximum discrepancy between the
two data sets is observed at PA = 230 # -310 # , so it is perhaps
the same deviation from a pure circular rotation that is seen
at the bottom plot.
there is a deficiency of the velocity at PA < 280 # and an
excess at PA > 280 # . We think that just this e#ect has
been observed by Pismis et al. (1995) in their long­slit
cross­sections. We now have a full two­dimensional pic­
ture, though, and it becomes clear that this cannot be
a radial gas outflow; it may be either counterrotating gas
Fig. 7. Isophote characteristics together with the orientations
of the kinematical major axes for the stars and ionized gas in
the center of NGC 5055. The line of nodes determined from the
outermost isophote orientation is PA = 105 # and the galaxy
plane inclination of 55 # corresponds to (1 - b/a)0 = 0.43.
streaming around a thick minibar roughly aligned with the
line of nodes, or, more probably, a gas polar arc (ring?)
shifted from the center and perhaps wrapped around the
same minibar.
It remains unclear if there is really a minibar in the
center of NGC 5055. On the one hand, the morphology
of the Mgb surface distribution, the presence of the H II
spot at the major axis, and the kinematical disturbances
described above constitute evidence for it. On the other
hand, we find that the orientations of the kinematical and
photometric major axes in the center of NGC 5055 (Fig. 7)

468 V. L. Afanasiev and O. K. Sil'chenko: NGC 5055
are coincident. Both may deviate slightly from the line of
nodes of the outer disk, PA 0 = 105 # , but they do it to­
gether. Only one possibility for a bar remains under such
circumstances: if it is fully aligned with the line of nodes
of the circumnuclear rotation plane. But in such a con­
figuration the velocity profile along the major axis must
have a plateau in the center and it does not have one. The
problem remains to be solved.
5. Discussion
The kinematical and stellar population parameter distri­
butions in the center of NGC 5055 have been shown to
be so complex that they cannot yet be interpreted unam­
biguously. Among a dozen spiral galaxies with chemically
distinct nuclei studied by us with the MPFS, we have not
seen any analogous cases. In the absence of the obser­
vational counterparts, a good choice would be to sketch
an environment of this complex central structure and to
try to understand what should proceed inevitably in such
environment from the dynamical point of view. But as
we mentioned in the Introduction, the whole structure
of NGC 5055 is highly ambiguous. Our own attempts to
clarify it have not been very productive. The brightness
profile in the range of radius of 2 ## -12 ## derived from the
NICMOS/HST data is well­fitted by a de Vaucouleurs' for­
mula indicating a classical bulge. But in the same range
of radius the ionized gas spirals are seen -- mainly to the
west of the nucleus (Boker et al. 1999); therefore for some
reason the prominent bulge is not able to stabilize the in­
ner disk. Could the bulge be non­axisymmetric? Its major
axis is aligned in PA # 95 # so being turned by 10 # with
respect to the line of nodes, and the ellipticity is constant
but low (Fig. 7). But the kinematical major axes of the
stars and ionized gas trace the photometric major axis
with a high precision instead of being turned by the same
10 # in opposite sense (Monnet et al. 1992); so they rotate
axisymmetrically. Is it due to a possible strong mass con­
centration in the very center? All these questions remain
still unclarified. Now we can only imagine some possible
configurations in the center of NGC 5055 and give a gen­
eral qualitative description.
1. A lopsided circumnuclear stellar disk.
A wave perturbation with an m = 1 has now become a
popular field of consideration. Theoreticians predicted
one­sided bars (e.g. Colin & Athanassoula 1989) and
eccentric nuclei (e.g. Miller & Smith 1992) some time
ago, one­armed spirals were also favoured by some dy­
namical models. Recently a growing amount of obser­
vational data has begun to confirm the reality of such
structures, in particular in the centers of normal spi­
ral galaxies. The most famous lopsided circumnuclear
stellar disk belongs to M 31: its two brightness centers,
one with high stellar velocity dispersion and the other
dynamically cold, have been explained by Tremaine
(1995) as an eccentric Keplerian disk around a su­
permassive black hole, and Bacon et al. (2001) have
argued that this disk must precess with an angular
velocity of some 3 km s -1 pc -1 so su#ering an m = 1
mode. In NGC 5055 there are too many asymmetries
along the major axis: the brightness asymmetry, in­
cluding [N II] and H# emission lines, the rotation and
stellar velocity dispersion asymmetries, and finally, the
Mgb index distribution asymmetry found by us in this
work. Whereas a sole brightness asymmetry can always
be explained by the dust projection e#ect (though it
is usually more pronounced along the minor axis), the
whole complex of asymmetries seen in the center of
NGC 5055 proves its intrinsic physical reality; in par­
ticular, the dust cannot a#ect narrow­band spectral
features, such as the Mgb index, and cannot create ar­
tificial Mgb enhancement to the south­west from the
center. We can imagine a chemically distinct, rather
young circumnuclear stellar disk traced by the high
Mgb index; it can precess slowly so that stars born, say,
a few Gyr ago at the eastern circumnuclear H II region
are now to the west of the dynamical center. We know
that there is a mass concentration in the nucleus of
NGC 5055 because of the rotation velocity peak near
the center; it may be a compact young stellar cluster
(Maoz et al. 1995) or a supermassive black hole, so the
whole situation may be similar to that in M 31, though
at a larger scale.
2. Another possibility which always helps to resolve a
complex situation is a minor merger. We could iden­
tify the magnesium­enhanced site with the core of the
merged galaxy and call it ``a secondary nucleus''. But
in this particular case the hypothesis of minor merger
meets a lot of problems. The fact of the exact coin­
cidence of the mean ages of stellar populations in the
primary and secondary nuclei looks somewhat strange.
Did the nuclei host synchronous star formation bursts
prior to merging? Moreover, the ``secondary'' nucleus
has a metallicity twice as high as that of the ``primary''
nucleus; more metal­rich nuclei belong usually to more
luminous and massive galaxies; if so, the merger could
not be a ``minor'' one. And finally, as we see a clear
azimuthal asymmetry of the merger remnants distri­
bution, the event should be very young, not older than
one orbital period, #10 7 yr, and then we should see
some other signatures of merging: tidal tails, global
disk heating and disappearance of spiral structure, etc.
We see none.
So the first hypothesis seems to be preferable.
Acknowledgements. We thank the post­graduate student of
SAO RAS A. V. Moiseev for supporting the observations at
the 6 m telescope. The 6 m telescope is operated under the
financial support of Science Ministry of Russia (registration
number 01­43). During the data analysis we have used the
Lyon­Meudon Extragalactic Database (LEDA) supplied by the
LEDA team at the CRAL­Observatoire de Lyon (France) and
the NASA/IPAC Extragalactic Database (NED) which is op­
erated by the Jet Propulsion Laboratory, California Institute
of Technology, under contract with the National Aeronautics
and Space Administration. The research is partly based on ob­
servations made with the NASA/ESA Hubble Space Telescope,

V. L. Afanasiev and O. K. Sil'chenko: NGC 5055 469
obtained from the data archive at the Space Telescope Science
Institute, which is operated by the Association of Universities
for Research in Astronomy, Inc., under NASA contract
NAS 5­26555. The work was supported by the grant 1.2.4.1 of
the Russian State Scientific­Technical Program ``Astronomy.
Basic Space Researches'' (the ``Astronomy'' section).
References
Acaretta, J. R., Manteiga, M., Pismis, P., Mampaso, A., &
Cruz­Gonzalez, G. 1996, AJ, 112, 1894
Afanasiev, V. L., Vlasyuk, V. V., Dodonov, S. N., & Sil'chenko,
O. K. 1990, Preprint SAO N54, Nizhnij Arkhyz: Special
Astrophys. Obs.
Afanasiev, V. L., Dodonov, S. N., Drabek, S. V., & Vlasyuk,
V. V. 1996, MPFS Manual. Nizhnij Arkhyz: SAO Publ.
Bacon, R., Emsellem, E., Combes, F., et al. 2001, A&A, 371,
409
Baggett, W. E., Baggett, S. M., & Anderson, K. S. J. 1998,
AJ, 116, 1626
Boker, T., Calzetti, D., Sparks, W., et al. 1999, ApJS, 124, 95
Bosma, A. 1981, AJ, 86, 1791
Burbidge, E. M., Burbidge, G. R., & Prendergast, K. H. 1960,
ApJ, 131, 282
Cardiel, N., Gorgas, J., Cenarro, J., & Gonzalez, J. J. 1998,
A&AS, 127, 597
Colin, J., & Athanassoula, E. 1989, A&A, 214, 99
Elmegreen, D. M. 1981, ApJS, 47, 229
Elmegreen, D. M., & Elmegreen, B. G. 1987, ApJ, 314, 3
Emsellem, E., Bacon, R., Monnet, G., & Poulain, P. 1996,
A&A, 312, 777
Fillmore, J. A., Boroson, T. A., & Dressler, A. 1986, ApJ, 302,
208
Fish, R. A. 1961, ApJ, 134, 880
Frei, Z., Guhathakurta, P., Gunn, J. E., & Tyson, J. A. 1996,
AJ, 111, 174
Heraudeau, P., & Simien, F. 1996, A&AS, 118, 111
Heraudeau, P., & Simien, F. 1998, A&AS, 133, 317
Kent, S. M. 1987, AJ, 93, 816
Maoz, D., Filippenko, A. V., Ho, L. C., et al. 1995, ApJ, 440, 91
Miller, R. H., & Smith, B. F. 1992, ApJ, 393, 508
Moiseev, A. V., & Mustsevoy, V. V. 2000, Pis'ma v AZh, 26,
657
Monnet, G., Bacon, R., & Emsellem, E. 1992, A&A, 253, 366
Pismis, P., Mampaso, A., Manteiga, M., Recillas, E., &
Cruz­Gonzalez, G. 1995, AJ, 109, 140
Pogge, R. W. 1989, ApJS, 71, 433
Sil'chenko, O. K. 1994, AZh, 71, 706
Sil'chenko, O. K. 1999a, AJ, 117, 2725
Sil'chenko, O. K. 1999b, AJ, 118, 186
Sil'chenko, O. K., & Afanasiev, V. L. 2000, A&A, 364, 479
Stasinska, G., & Sodre, Jr. I. 2001, A&A, 374, 919
Tantalo, R., Chiosi, C., & Bressan, A. 1998, A&A, 333, 419
Thornley, M. D. 1996, ApJ, 469, L45
Thornley, M. D., & Mundy, L. G. 1997, ApJ, 484, 202
Tremaine, S. 1995, AJ, 110, 628
Vlasyuk, V. V. 1993, Astrofiz. issled. (Izv. SAO RAS) 36, 107
Worthey, G. 1994, ApJS, 95, 107
Worthey, G., Faber, S. M., & Gonzalez, J. J. 1992, ApJ, 398, 69
Worthey, G., Faber, S. M., Gonzalez, J. J., & Burstein, D.
1994, ApJS, 94, 687